JP2018151805A - Data item name estimating apparatus, data item name estimating method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a data item name estimating apparatus, a data item name estimating method, and a program capable of estimating an item name given to a data item without requiring knowledge concerning data registered in a database.SOLUTION: A data item name estimating apparatus 100 comprises: a learning processing unit 10 which extracts feature values of data to which a data item name is assigned for each data item name in a learning table, defines respective relationships between the extracted feature values and attributes corresponding to the respective data item names, specifies the corresponding data item name for each attribute, learns combinations of the attributes and the respective data item names in the learning table to create a learning model; and an estimation processor 20 which compares the feature values of each data item of the target table with the definition by the learning processing unit 10 to estimate attributes of each data item, and applies the estimated attributes of each data item to the learning model to estimate the data item name of the data item of the target table.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a data item name estimation device and a data item name estimation method for estimating a data item name of a table whose data item name is unknown, and further relates to a program for realizing these.

  In recent years, with the development of ICT (Information and Communication Technology), it has become possible and easy to acquire a large amount of information in real time, and a large amount of information has been collected and accumulated. When various analyzes and predictions are performed using these pieces of information, it is necessary to create a database having a common schema by integrating a plurality of databases storing these pieces of information.

  For database integration, it is necessary to define common classifications or attributes. For this reason, conventionally, a database or data modeling specialist confirms the schema configuration with each database administrator and manually identifies, extracts, and integrates common data. Therefore, it takes a lot of work to integrate the databases.

  Further, when performing data integration, it is easy to determine which items should be integrated if the person who performs the work understands the schema structure for each data to be integrated. However, if the person who performs the work does not have sufficient knowledge about the data to be integrated and a meaningless item name such as “t1” is defined as the data item name, which item should be integrated? It is very difficult to judge whether it is good. Therefore, from this point of view, database integration is a time-consuming operation.

  In order to solve such a problem, for example, Patent Literature 1 discloses a classification construction support system. The classification construction support system disclosed in Patent Document 1 first extracts a feature value of a data item by using record data of the data item. Next, the classification construction support system classifies each of the plurality of data items into an appropriate classification or attribute by obtaining the similarity with the attribute defined in advance for the extracted feature value.

JP 2006-99236 A

  However, in the classification construction support system disclosed in Patent Document 1, data items having the same feature value can be combined into one attribute, but it is impossible to determine the name of the data item having the attribute. is there.

  For example, it is assumed that “temperature” is determined as an attribute for classifying a certain data item. In this case, in order to perform data integration, it is appropriate that the data item having the attribute of “temperature” defines any item name of “average temperature”, “maximum temperature”, and “minimum temperature”. It was difficult to identify.

  An example of the object of the present invention is to solve the above-mentioned problems and to estimate an item name given to a data item without requiring knowledge about data registered in a database. An object is to provide an item name estimation method and a program.

In order to achieve the above object, a data item name estimation apparatus according to one aspect of the present invention includes:
For each data item name in the learning table, the feature value of the data to which the data item name is assigned is extracted, the relationship between the extracted feature value and the attribute corresponding to the data item name is defined, and the attribute A learning processing unit for identifying a corresponding data item name for each, learning a combination of the attribute and the data item name in the learning table, and creating a learning model;
The feature value for each data item of the target table is collated with the definition by the learning processing unit, the attribute of each data item is estimated, the attribute for each estimated data item is applied to the learning model, An estimation processing unit that estimates a data item name of a data item of the target table;
It is characterized by having.

In order to achieve the above object, a data item name estimation method according to an aspect of the present invention includes:
(A) For each data item name in the learning table, the feature value of the data to which the data item name is assigned is extracted, and the relationship between the extracted feature value and the attribute corresponding to the data item name is defined. Identifying a corresponding data item name for each attribute, learning a combination of the attribute and the data item name in the learning table, and creating a learning model;
(B) The feature value for each data item of the target table is collated with the definition obtained in the step (a), the attribute of each data item is estimated, and the attribute for each estimated data item is Applying to the learning model and estimating a data item name of a data item of the target table;
It is characterized by having.

Furthermore, in order to achieve the above object, a program according to one aspect of the present invention is provided.
On the computer,
(A) For each data item name in the learning table, the feature value of the data to which the data item name is assigned is extracted, and the relationship between the extracted feature value and the attribute corresponding to the data item name is defined. Identifying a corresponding data item name for each attribute, learning a combination of the attribute and the data item name in the learning table, and creating a learning model;
(B) The feature value for each data item of the target table is collated with the definition obtained in the step (a), the attribute of each data item is estimated, and the attribute for each estimated data item is Applying to the learning model and estimating a data item name of a data item of the target table;
Is executed.

  As described above, according to the present invention, it is possible to estimate an item name given to a data item without requiring knowledge about data registered in the database.

FIG. 1 is a block diagram showing a schematic configuration of a data item estimation device according to an embodiment of the present invention. FIG. 2 is a block diagram showing a specific configuration of the data item estimation apparatus according to the embodiment of the present invention. FIG. 3 is a diagram showing an example of a learning table used in the embodiment of the present invention. FIG. 4 is a diagram showing an example of a target table that is a target of data item name estimation processing in the embodiment of the present invention. FIG. 5 is a flowchart showing the operation during the learning process of the data item name estimation apparatus according to the embodiment of the present invention. FIG. 6 is a diagram showing an example of attribute information created in the embodiment of the present invention. FIG. 7 is a diagram showing an example of attribute item information created in the embodiment of the present invention. FIG. 8 is a diagram showing an example of attribute combination information created in the embodiment of the present invention. FIG. 9 is a flowchart showing an operation during the estimation process of the data item name estimation apparatus according to the embodiment of the present invention. FIG. 10A is a diagram showing an example of feature values extracted from the target table used in the embodiment of the present invention. FIG. 10B is a diagram illustrating an example of the similarity calculated from the feature values illustrated in FIG. FIG. 11A is a diagram showing an example of attribute combination information used in the embodiment of the present invention. FIG.11 (b) is a figure which shows an example of the appearance frequency of the data item name calculated in embodiment of this invention. FIG. 12 is a block diagram illustrating an example of a computer that implements the data item estimation apparatus according to the embodiment of the present invention.

(Summary of Invention)
In the present invention, the feature value is extracted from the data item of the table that is the estimation target of the data item name, the similarity between the extracted feature value and the attribute is calculated, and each data item is assigned to the attribute having a high similarity. Classify. Then, by applying the combination of data item attributes that make up the estimation target table to the correspondence between the data item attribute combination and the data item name learned in advance, The correct data item name is given. That is, in the present invention, learning processing and estimation processing are mainly performed. Specific description will be given below.

(Embodiment)
Hereinafter, a data item estimation device, a data item estimation method, and a program according to an embodiment of the present invention will be described with reference to FIGS.

[Device configuration]
Initially, the structure of the data item estimation apparatus in this Embodiment is demonstrated. FIG. 1 is a block diagram showing a schematic configuration of a data item estimation device according to an embodiment of the present invention.

  A data item estimation device 100 according to the present embodiment shown in FIG. 1 is a device that estimates a data item name of a table to be processed (hereinafter referred to as “target table”). As shown in FIG. 1, the data item estimation device 100 includes a learning processing unit 10 and an estimation processing unit 20.

  First, the learning processing unit 10 extracts, for each data item name in the learning table, the feature value of the data to which the data item name is assigned, and the extracted feature value and the attribute corresponding to the data item name Define the relationship. Next, the learning processing unit 10 specifies a corresponding data item name for each attribute, learns the frequency of appearance of the attribute and the data item name in the learning table, and creates a learning model.

  First, the estimation processing unit 20 collates the feature value for each data item in the target table with the definition by the learning processing unit 10 to estimate the attribute of each data item. Next, the estimation processing unit 20 applies the estimated attribute for each data item to the learning model, and estimates the data item name of the data item in the target table.

  As described above, in the present embodiment, a learning model for estimating a data item name is generated using a learning table with a known data item name. By using this learning model, the data item name is Data item names in unknown tables are inferred. For this reason, according to this Embodiment, the item name provided to a data item can be estimated, without requiring the knowledge regarding the data registered into the database.

  Next, the configuration of the data item estimation apparatus 100 in the present embodiment will be described more specifically with reference to FIGS. FIG. 2 is a block diagram showing a specific configuration of the data item estimation apparatus according to the embodiment of the present invention. FIG. 3 is a diagram showing an example of a learning table used in the embodiment of the present invention. FIG. 4 is a diagram showing an example of a target table that is a target of data item name estimation processing in the embodiment of the present invention.

  As shown in FIG. 2, in the present embodiment, the data item estimation device 100 includes a storage unit 30 in addition to the learning processing unit 10 and the estimation processing unit 20. The storage unit 30 stores attribute information 31, attribute individual information 32, and attribute combination information 33, which will be described later.

  As shown in FIG. 2, the learning processing unit 10 includes a learning table receiving unit 11, a feature extracting unit 12, an attribute information creating unit 13, an attribute individual information creating unit 14, and an attribute combination information creating unit 15. And.

  The learning table reception unit 11 receives a learning table (see FIG. 3) input from the outside, and passes the received learning table to the feature extraction unit 12. The learning table is data in a table format as shown in FIG. 3. The format of each record data is any of XML (Extensible markup language), CSV (Common Separated Value), HTML (Hypertext markup language), etc. It may be in the form of

  For each data item name in the learning table, the feature extraction unit 12 extracts a feature value from the data of the record to which the data item name is assigned. The extracted feature values include data type (character type, numerical type), statistical information (average value, variance value), appearance frequency of the same record, and the like (see FIG. 6 described later). Moreover, as a method for extracting feature values from the table, the method disclosed in Patent Document 1 described above can be cited.

  The attribute information creation unit 13 creates attribute information 31 (see FIG. 6 described later) that defines the relationship between the feature value extracted by the feature extraction unit 12 and the attribute corresponding to the data item name of the extraction source data. The created attribute information is stored in the storage unit 30.

  Specifically, the attribute information creation unit 13 sets the feature value extracted by the feature extraction unit 12 from the outside when an attribute corresponding to the data item name of the extraction source data is set. Attribute information 31 is created for associating the attribute. Further, the attribute setting may be performed manually.

  The attribute individual information creation unit 14 creates attribute individual information (see FIG. 7) to which a corresponding data item name is assigned for each attribute included in the attribute information created by the attribute information creation unit 13.

  Specifically, when the corresponding data item name is set for each attribute included in the attribute information, the attribute individual information creation unit 14 assigns the attribute and the input data item name. Attribute individual object information is created using objects. The attribute individual information indicates the correspondence between attributes and individual objects. Here, the individual item means an object to which a data item name is assigned. The data item name may be set manually.

  The attribute combination information creation unit 15 creates attribute combination information (see FIG. 8) indicating the combination of attributes in the learning table and the corresponding data item name as a learning model using the attribute individual object information. .

  As shown in FIG. 2, the estimation processing unit 20 includes a target table reception unit 21, a feature extraction unit 22, an attribute estimation unit 23, an item name estimation unit 24, an estimation result display unit 25, and a result edit. Part 26.

  The target table reception unit 21 receives a target table (see FIG. 4) input from the outside, and passes the received target table to the feature extraction unit 22. Similar to the learning table, the target table is also data in a table format as shown in FIG. Also, the format of each record data in the target table may be any format such as XML (Extensible markup language), CSV (Common Separated Value), HTML (Hypertext markup language).

  The feature extraction unit 22 extracts the feature value of the data for each data item from the record of the target table. The extracted feature values include data type (character type, numerical type), statistical information (average value, variance value), appearance frequency of the same record, and the like. Moreover, as a method for extracting feature values from the table, the method disclosed in Patent Document 1 described above can be cited.

  The attribute estimation unit 23 collates the feature value extracted by the feature extraction unit 22 with the attribute information 31 to estimate the attribute of each data item. Specifically, the attribute estimation unit 23 first calculates the similarity by comparing the feature value of each data item with the feature value of each attribute included in the attribute information 31. And the attribute estimation part 23 specifies the attribute with the highest similarity with the feature value of the data item used as an estimation object, and estimates the specified attribute as an attribute of the data item used as an estimation object ( (Refer FIG. 10 (a) and (b) mentioned later).

  The item name estimation unit 24 first sets a combination of attributes using the attributes estimated by the attribute estimation unit 23. Next, the item name estimation unit 24 calculates the frequency of occurrence of a specific data item name using the attribute combination information 33 and the attribute individual information 32 for each set attribute combination. And the item name estimation part 24 estimates the data item name of each data item of a target table based on a calculation result.

  Specifically, the item name estimation unit 24 uses the attribute individual information 32 to identify one or more data item names that may correspond to each attribute for which a combination is set. Then, using the attribute combination information 33, the item name estimation unit 24 calculates the appearance frequency (appearance probability) for each data item name that may correspond to each set attribute combination. And the item name estimation part 24 specifies the data item name with the highest appearance frequency among the data item names which may correspond as a data item name of the estimated attribute.

  The estimation result display unit 25 displays the data item name estimated by the item name estimation unit 24 on the screen of a display device connected to the data item name estimation device 100 or a display device of an external terminal device. The result editing unit 26 corrects the estimation result by the item name estimation unit 24 when it is determined that there is an error in the estimated data item name and the corrected data item name is input from the outside. Further, the result editing unit 26 can transmit the correction content to the learning processing unit 10. In this case, the learning processing unit 10 reflects the correction contents in the attribute information 31, the attribute individual information 32, and the attribute combination information 33.

[Device operation]
Next, the operation of the data item estimation apparatus 100 in the present embodiment will be described with reference to FIGS. Moreover, in this Embodiment, the data item estimation method is implemented by operating the data item estimation apparatus 100. Therefore, the description of the data item estimation method in the present embodiment is replaced with the following description of the operation of the data item estimation apparatus 100.

  First, the learning phase (learning process) will be described with reference to FIGS. FIG. 5 is a flowchart showing the operation during the learning process of the data item name estimation apparatus according to the embodiment of the present invention.

  As shown in FIG. 5, first, the learning table receiving unit 11 receives a learning table (see FIG. 3) input from the outside (step A1). The learning table reception unit 11 passes the received learning table to the feature extraction unit 12.

  Next, the feature extraction unit 12 selects one of the data items in the accepted learning table, and extracts a feature value from the record data of the selected data item (step A2).

  Next, the feature extraction unit 12 determines whether or not feature values have been extracted from all data items (step A3). If the feature value is not extracted from all the data items as a result of the determination, the feature extracting unit 12 executes Step A2 again. If the feature value is extracted from all the data items, the attribute information is created. The unit 13 is instructed to execute step A4.

  If it is determined Yes in step A3, the attribute information creation unit 13 creates attribute information 31 by associating the set attribute for each data item (step A4). Further, in step A4, the attribute is set by the outside based on the data item name of the extraction source data. FIG. 6 is a diagram showing an example of attribute information created in the embodiment of the present invention.

  Next, the attribute individual information creation unit 14 creates attribute individual information 32 to which a corresponding data item name is assigned for each attribute included in the attribute information created in step A4 (step A5). In step A5, the data item name is set manually by the outside. FIG. 7 is a diagram showing an example of attribute item information created in the embodiment of the present invention.

  Next, the attribute combination information creation unit 15 uses the attribute individual information created in step A5 as the learning model to indicate the attribute combination information in the learning table and the corresponding data item name as the learning model. 33 is created (step A6). FIG. 8 is a diagram showing an example of attribute combination information created in the embodiment of the present invention.

  Next, an estimation phase (estimation process) will be described with reference to FIGS. FIG. 9 is a flowchart showing an operation during the estimation process of the data item name estimation apparatus according to the embodiment of the present invention.

  As shown in FIG. 9, first, the target table receiving unit 21 receives a target table (see FIG. 4) input from the outside (step B1). The target table receiving unit 21 passes the received target table to the feature extracting unit 22.

  Next, as shown in FIG. 10A, the feature extraction unit 22 selects one of the received data items of the target table, and extracts a feature value from the record data of the selected data item (step B2).

  FIG. 10A is a diagram showing an example of feature values extracted from the target table used in the embodiment of the present invention. FIG. 10B is a diagram illustrating an example of the similarity calculated from the feature values illustrated in FIG.

  Next, the feature extraction unit 22 determines whether or not feature values have been extracted from all data items (step B3). As a result of the determination, if the feature values are not extracted from all the data items, the feature extraction unit 22 executes Step B2 again, and if the feature values are extracted from all the data items, the attribute estimation unit. 23 is instructed to execute step B4.

  If the attribute estimation unit 23 determines Yes in step B3, the attribute estimation unit 23 selects one data item in the target table. Then, the attribute estimation unit 23 compares the selected data item feature value with the feature value of each attribute included in the attribute information 31 as shown in FIG. The attribute with the highest degree is estimated as the attribute of the data item (step B4).

  Next, the attribute estimation part 23 determines whether the attribute was estimated about all the data items (step B5). If the attribute is not estimated for all data items as a result of the determination, the attribute estimation unit 23 executes Step B4 again. If the attribute is estimated for all data items, the item name estimation unit 24 Is instructed to execute step B6.

  Next, when it is determined Yes in step B5, the attribute estimation unit 23 first sets a combination of a plurality of attributes that appear at the same time using the attribute estimated in step B4. Next, as shown in FIGS. 11A and 11B, the item name estimation unit 24 uses the attribute combination information 33 and the attribute individual information 32 for each set attribute combination to specify a specific data item name. Is calculated (appearance probability) (step B6).

  FIG. 11A is a diagram showing an example of attribute combination information used in the embodiment of the present invention. FIG.11 (b) is a figure which shows an example of the appearance frequency of the data item name calculated in embodiment of this invention. In FIG. 11B, “estimation target attribute: [attribute combination]” means an attribute that is an estimation target of a data item name in a certain attribute combination. For example, “date: [place name, date]” means that when the combination of attributes is “place name, date”, the estimation target of the data item name is “date”.

  In the present embodiment, the item name estimation unit 24 creates, for example, an association rule that estimates the data item name of a data item having a certain attribute from a combination of attributes. Further, the item name estimation unit 24 refers to the attribute individual information 32, identifies an individual (data item name) having the same attribute as the estimated attribute, and uses the association rule for the identified individual to obtain data. The probability of being selected as an item name is calculated.

Furthermore, the creation of an association rule can be performed by using, for example, an algorithm using the correlation disclosed in Reference Document 1 or Reference Document 2 below.
(Reference 1) G. Piatetsky-Shapiro (1991). Discovery, analysis, and presentation of strong rules. In G. Piatetsky-Shapiro and WJ Frawley, editors, Knowledge Discovery in Databases. AAAI / MIT Press, Cambridge, MA.
(Reference 2) R. Agrawal, T. Imielinski, and A. Swami (1993). Mining association rules between sets of items in large databases. In Proceedings of the 1993 ACM SIGMOD International Conference on Management of Data, pages 207-216 .

  Next, the item name estimation unit 24 selects one data item of the target table. Then, the item name estimation unit 24 collates the combination of the estimated attribute of the selected data item and the other attribute appearing at the same time with the calculation result obtained in step B6, and has the highest appearance probability. A high data item name is estimated as the data item name of the selected data item (step B7).

  Next, the item name estimation unit 24 determines whether data item names have been estimated for all data items (step B8). As a result of the determination, if the data item names are not estimated for all the data items, the item name estimation unit 24 executes Step B7 again.

  On the other hand, if it is determined that the data item names are estimated for all data items, the item name estimation unit 24 causes the estimation result display unit 25 to display the results. Thereby, the process in the data item name estimation apparatus 100 is complete | finished.

  In this embodiment, the item name estimation unit 24 receives the attribute combination information 33 as an input, forms a neural network that outputs the data item name, and estimates the data item name using the formed neural network. Also good. Further, the item name estimation unit 24 receives the attribute combination information 33 as input, constructs a Bayesian estimation for estimating the data item name having the highest appearance rate from the combination of the attributes of the data items, and uses the constructed Bayesian estimation. The data item name may be estimated.

  As described above, in the present embodiment, it is possible to estimate an item name given to a data item without requiring knowledge about data registered in the database. For this reason, if this embodiment is used, when data analysis or data integration is performed, data items having the same meaning and that can be integrated can be easily specified without knowledge of data.

[program]
The program in the present embodiment may be a program that causes a computer to execute steps A1 to A6 shown in FIG. 5 and steps B1 to B8 shown in FIG. By installing and executing this program on a computer, the data item estimation device 100 and the data item estimation method in the present embodiment can be realized. In this case, a CPU (Central Processing Unit) of the computer functions as the learning processing unit 10 and the estimation processing unit 20 and performs processing.

  Further, in the present embodiment, the storage unit 30 stores data files constituting the attribute information 31, the attribute individual information 32, and the attribute combination information 33 in a storage device such as a hard disk provided in the computer. realizable.

  The program in the present embodiment may be executed by a computer system constructed by a plurality of computers. In this case, for example, each computer may function as either the learning processing unit 10 or the estimation processing unit 20. The storage unit 30 may be constructed on a computer different from the computer that executes the program in the present embodiment.

  Here, a computer that realizes the data item estimation apparatus 100 by executing the program according to the present embodiment will be described with reference to FIG. FIG. 12 is a block diagram illustrating an example of a computer that implements the data item estimation apparatus according to the embodiment of the present invention.

  As shown in FIG. 12, the computer 110 includes a CPU 111, a main memory 112, a storage device 113, an input interface 114, a display controller 115, a data reader / writer 116, and a communication interface 117. These units are connected to each other via a bus 121 so that data communication is possible.

  The CPU 111 performs various calculations by developing the program (code) in the present embodiment stored in the storage device 113 in the main memory 112 and executing them in a predetermined order. The main memory 112 is typically a volatile storage device such as a DRAM (Dynamic Random Access Memory). Further, the program in the present embodiment is provided in a state of being stored in a computer-readable recording medium 120. Note that the program in the present embodiment may be distributed on the Internet connected via the communication interface 117.

  Specific examples of the storage device 113 include a hard disk drive and a semiconductor storage device such as a flash memory. The input interface 114 mediates data transmission between the CPU 111 and an input device 118 such as a keyboard and a mouse. The display controller 115 is connected to the display device 119 and controls display on the display device 119.

  The data reader / writer 116 mediates data transmission between the CPU 111 and the recording medium 120, and reads a program from the recording medium 120 and writes a processing result in the computer 110 to the recording medium 120. The communication interface 117 mediates data transmission between the CPU 111 and another computer.

  Specific examples of the recording medium 120 include general-purpose semiconductor storage devices such as CF (Compact Flash (registered trademark)) and SD (Secure Digital), magnetic recording media such as a flexible disk, or CD- An optical recording medium such as ROM (Compact Disk Read Only Memory) can be used.

  Note that the data item estimation apparatus 100 according to the present embodiment can be realized not by using a computer in which a program is installed but also by using hardware corresponding to each unit. Furthermore, the data item estimation apparatus 100 may be partially realized by a program and the remaining part may be realized by hardware.

  When performing data integration, it is possible to estimate a data structure having the same data item by estimation even if the data to be integrated does not have knowledge of the meaning of the data structure and the data item.

DESCRIPTION OF SYMBOLS 10 Learning processing part 11 Learning table reception part 12 Feature extraction part 13 Attribute information creation part 14 Attribute piece information creation part 15 Attribute combination information creation part 20 Estimation processing part 21 Target table reception part 22 Feature extraction part 23 Attribute estimation part 24 Item Name estimation unit 25 Estimation result display unit 26 Result editing unit 30 Storage unit 31 Attribute information 32 Attribute item information 33 Attribute combination information 100 Data item name estimation device 110 Computer 111 CPU
112 Main Memory 113 Storage Device 114 Input Interface 115 Display Controller 116 Data Reader / Writer 117 Communication Interface 118 Input Device 119 Display Device 120 Recording Medium 121 Bus

Claims (9)

For each data item name in the learning table, the feature value of the data to which the data item name is assigned is extracted, the relationship between the extracted feature value and the attribute corresponding to the data item name is defined, and the attribute A learning processing unit for identifying a corresponding data item name for each, learning a combination of the attribute and the data item name in the learning table, and creating a learning model;
The feature value for each data item of the target table is collated with the definition by the learning processing unit, the attribute of each data item is estimated, the attribute for each estimated data item is applied to the learning model, An estimation processing unit that estimates a data item name of a data item of the target table;
A data item name estimation device comprising: The learning processing unit
Create attribute information that defines the relationship between the extracted feature value and the attribute corresponding to the data item name,
Further, for each attribute included in the created attribute information, create attribute individual information with a corresponding data item name,
Then, using the attribute individual information, as the learning model, to create attribute combination information indicating a combination of the attributes in the learning table and a data item name corresponding to each of the attributes,
The data item name estimation apparatus according to claim 1. The estimation processing unit
Extract feature values for each data item from the records in the target table,
The extracted feature value is collated with the attribute information to estimate the attribute of each data item,
Then, the attribute combination is set using the estimated attribute, and the frequency of occurrence of a specific data item name is calculated for each set attribute combination using the attribute combination information and the attribute individual information. ,
Based on the calculation result, estimate the data item name of each data item of the target table,
The data item name estimation device according to claim 2. (A) For each data item name in the learning table, the feature value of the data to which the data item name is assigned is extracted, and the relationship between the extracted feature value and the attribute corresponding to the data item name is defined. Identifying a corresponding data item name for each attribute, learning a combination of the attribute and the data item name in the learning table, and creating a learning model;
(B) The feature value for each data item of the target table is collated with the definition obtained in the step (a), the attribute of each data item is estimated, and the attribute for each estimated data item is Applying to the learning model and estimating a data item name of a data item of the target table;
A data item name estimation method characterized by comprising: In the step (a),
Create attribute information that defines the relationship between the extracted feature value and the attribute corresponding to the data item name,
Further, for each attribute included in the created attribute information, create attribute individual information with a corresponding data item name,
Then, using the attribute individual information, as the learning model, to create attribute combination information indicating a combination of the attributes in the learning table and a data item name corresponding to each of the attributes,
The data item name estimation method according to claim 4. In the step (b),
Extract feature values for each data item from the records in the target table,
The extracted feature value is collated with the attribute information to estimate the attribute of each data item,
Then, the attribute combination is set using the estimated attribute, and the frequency of occurrence of a specific data item name is calculated for each set attribute combination using the attribute combination information and the attribute individual information. ,
Based on the calculation result, estimate the data item name of each data item of the target table,
The data item name estimation method according to claim 5. On the computer,
(A) For each data item name in the learning table, the feature value of the data to which the data item name is assigned is extracted, and the relationship between the extracted feature value and the attribute corresponding to the data item name is defined. Identifying a corresponding data item name for each attribute, learning a combination of the attribute and the data item name in the learning table, and creating a learning model;
(B) The feature value for each data item of the target table is collated with the definition obtained in the step (a), the attribute of each data item is estimated, and the attribute for each estimated data item is Applying to the learning model and estimating a data item name of a data item of the target table;
A program that executes In the step (a),
Create attribute information that defines the relationship between the extracted feature value and the attribute corresponding to the data item name,
Further, for each attribute included in the created attribute information, create attribute individual information with a corresponding data item name,
Then, using the attribute individual information, as the learning model, to create attribute combination information indicating a combination of the attributes in the learning table and a data item name corresponding to each of the attributes,
The program according to claim 7. In the step (b),
Extract feature values for each data item from the records in the target table,
The extracted feature value is collated with the attribute information to estimate the attribute of each data item,
Then, the attribute combination is set using the estimated attribute, and the frequency of occurrence of a specific data item name is calculated for each set attribute combination using the attribute combination information and the attribute individual information. ,
Based on the calculation result, estimate the data item name of each data item of the target table,
The program according to claim 8.

Images